Nuneaton and Bedworth Borough Council v RH and Secretary of State for Work and Pensions (2017) UKUT 471 (AAC)
A very interesting (and perhaps surprising) Upper Tribunal 3 judge decision on the issue of ‘what is a bedroom’ for the purposes of Housing Benefit Reg 13 – the bedroom tax.
RH occupied property described as a 3 bedroom house. Under the tenancy agreement the maximum number of permitted occupiers was 4. RH and her husband used one room. Her two sons, 7 and 9 at the relevant time, occupied one room each. As under Reg 13, the two sons were expected to share a room – not ‘entitled’ to one each, the bedroom tax was applied. However, the FTT found as facts that:
neither of the smaller rooms could be shared. One of the two rooms could accommodate a single bed, but a bunk would obscure much of the natural light. That was not in dispute. The other room could accommodate a bunk bed without interfering with the doorway, natural light, and access to the room, but the child sleeping on the top bunk would be unacceptably close to the light fitting, even if it were changed to a flat rather than pendant fitting, and the bunks would not allow the wardrobe to be opened fully.
This was the basis if the FTT appeal and UT appeal. If the rooms could not accommodate both children, should the bedroom tax apply (the ‘actual occupiers’ or ‘connection’ issue).
The relevant bit of Reg B13 is as follows:
(5) The claimant is entitled to one bedroom for each of the following categories of person whom the relevant authority is satisfied occupies the claimant’s dwelling as their home (and each person shall come within the first category only which is applicable)-
(a) a couple (within the meaning of Part 7 of the Act);
(b) a person who is not a child;
(ba) a child who cannot share a bedroom;
(c) two children of the same sex;
(d) two children who are less than 10 years old;
(e) a child.
The DWP’s case was that this meant first identifying the occupants of the property by placing them in one of these categories, (this was generally agreed), and then identify the number of bedrooms, without reference to the categories of person who would have to occupy them. Thus a house with 3 small bedrooms that could only be occupied by a child under 10 would still have three bedrooms.
The Upper Tribunal disagreed.
First, to us that is not the natural meaning of the language of paragraph (5). On Mr Brown’s (DWP) approach, the paragraph sets up a calculation by reference to the actual occupants as classified into particular categories but then ignores the inevitable characteristics of the categories, such as that they consist of two people or people of a particular age. The paragraph provides that the claimant is entitled to a bedroom for each category. The natural expectation of that language is that the room would be a bedroom for the persons bearing the characteristics of that category, not a room that ignored those characteristics. This leads on to our second reason. If the legislation were to produce the result that Mr Brown and Ms Meacher (LA) contended for, it would need much clearer language to show that it was necessary to sever the claimant’s entitlement from the characteristics of the categories as set out in paragraph (5). The language does not do that.
In the circumstances, RH’s house meant that (5)(c) and (5)(d) could not apply as the rooms were not suited for people with those characteristics, so (5)(e) applied and they were entitled for a room for each child.
So far, so good. But then there are two previous judgments on this issue, which the Upper Tribunal then turns to: Secretary of State for Work and Pensions v Nelson and Fife Council (2014) UKUT 525 (AAC) (our report here) and Secretary of State for Work and Pensions v Glasgow City Council and IB (2017) CSIH 35 (our report here). Both of these decisions had touched the ‘connection’ issue, to different degrees of directness.
Nelson effectively found that any ‘bedroom’ that was sufficient to accommodate anyone on the B13(5) list was a bedroom.
IB held that the number of bedrooms should be determined by the property in its vacant state, without regard to “the actual needs of the occupants or how they use the rooms for whatever reason from time to time.”
The UT held it could diverge from Nelson – also a 3 judge UT decision:
First, the Upper Tribunal’s approach to precedent is not rigid. It recognises that there may be compelling reasons why even a single judge should not follow a decision by a three-judge panel; and a slavish adherence to the standard approach should not allow the perpetuation of error.
Second, we were appointed to consider the connection issue by our Chamber President, who presided in Nelson. He can only have done so in the knowledge that we might disagree with that decision.
Third, the connection issue did not arise in Nelson.
Fourth, although the panel mentioned the connection issue and, no doubt, considered it an important part of its analysis, the actual focus of the reasoning thereafter was on the classification issue and in particular the features and use questions. We can find nothing to indicate that the panel’s reasoning on those questions was affected by its view of the connection issue.
Perhaps more problematically, the decision in IB was a decision of the Scottish Inner Court of Session (arguably Court of Appeal equivalent). The UT distinguishes IB as follows:
we doubt whether we would have followed the reasoning in IB even if it had been a decision of the Court of Appeal in England and Wales. We say that because IB was not concerned with the connection issue. Rather, it was concerned with the classification issue and in particular with the change of use question. The room in question had been a bedroom, but it had been changed for use as a living room for the claimant. So it is not necessary for us to decide whether the Court was right on the issue before it. What we have to decide is whether we should apply the Court’s reasoning to the issue before us. We have decided not to do so. The Court did not consider the issue before us and, so far as the decision shows, the reasoning on which we rely was not put to the Court in argument. The Court did not consider that reasoning, let alone reject it.
However, to the extent that the IB decision did clearly impact on the UT’s decision here, the issue of whether the UT was bound by the Court of Session decision arose. The UT held:
Decisions of the Court of Appeal are binding on the Upper Tribunal. This does not mean that decisions of the Court of Session are also binding. Laws LJ explained why in Marshalls Clay Products Ltd v Caulfield, Clarke v Frank Staddon Ltd [2004] ICR 1502 at [32]:
The rules of precedent or stare decisis cognisable here are given by the common law . . . The essence is that precedent confines the very power of the courts subject to it. It is not a rule of discretion or comity or anything of the kind. It is therefore of necessity a doctrine whose reach is limited to the jurisdiction in which the courts in question operate. The House of Lords is no exception; by statute its writ runs to three jurisdictions, and accordingly it binds the lower courts within each of those jurisdictions. Statute might also extend the scope of precedent, as was done by the European Communities Act 1972 . . . Now, statutory provisions which give dominion to courts in one jurisdiction (international or otherwise) over courts in another are apt, here at least, to father constitutional tensions. But it is at least clear, and here is the point on this part of the case, that it would be a constitutional solecism of some magnitude to suggest that by force of the common law of precedent any court of England and Wales is in the strict sense bound by decisions of any court whose jurisdiction runs in Scotland only or – most assuredly – vice versa. Comity and practicality are another thing altogether. They exert a wholly legitimate pressure.
IB is persuasive authority in the Upper Tribunal when it exercises jurisdiction in England (as it is doing in this case), but no more.
This is perhaps questionable, not least as the UT has a single national jurisdiction.
Comment
While this is a sensible, practical decision, I strongly suspect it will be appealed, on both the decision and the precedent point. Which would present the prospect of the Court of Appeal ruling both on what amounts to a bedroom and on whether a Scottish Inner Court of Session decision was binding on the Upper Tribunal sitting in England.
I am more doubtful whether this case will be appealed, for three reasons: (1) On the facts as found by the First-tier Tribunal the Secretary of State’s case is pretty well unarguable; but (2) the Upper Tribunal sent the case back to the F-tT to look again at one aspect of the facts and I can’t see the Court of Appeal granting permission to appeal when the facts are unresolved. (3) On the general issue of whether the Inner House of the Court of Session binds the UT and F-tT in England and Wales, I can’t see the Court of Appeal being interested in departing from the principle set out by Laws LJ – they will be acutely aware of the implication in any finding that the Inner House binds tribunals in England and Wales that CA decisions bind tribunals in Scotland, and they know that would be anathema to the Scots.
There are circumstances where in effect a CS decision is binding in England and Wales, exemplified by a case called MMcK v SSWP last August. In that case the Secretary of State appealed a UT decision to the Inner House but lost comprehensively, so the UT decision, which is binding across the whole of Great Britain, stands, but now with added authority from an appellate court.
Take the point on the re-hearing, but any appeal would be on principle rather than the facts.
The position on whether the Court of Session is binding is not, I think quite as straightforward as the UT found. See Tehrani v Home Secretary [2007] AC 521 on the Immigration Appeals Tribunal and jurisdiction of High Court and Court of Session.
The comments on jurisdiction are extremely surprising, particularly in a decision which involved Judge Jacobs (who literally wrote the book on this sort of thing).
The problem is that there is no such thing as “the Upper Tribunal sitting in England” or the “Upper Tribunal sitting in Scotland”. There is only the Upper Tribunal. The Upper Tribunal is either bound by a decision or it is not. The physical location of the Judge cannot be the deciding factor, because it is open to the Judge to sit anywhere in the UK.
It seems to me that the House of Lords decision in Tehrani is still authority for the position that the UT is subject to the concurrent supervisory jurisdiction of the CoA, the EWHC and the CoS and, as a result, must follow case law from all of these sources as far as they are relevant.
I am amazed that everybody seems to have forgotten about Tehrani and that the UT in this case has quoted from an employment case pre-dating it as the only authority for its view.
Well this may all be angels dancing on a pinhead as in practice the appellate courts hardly ever seem to get into a dispute as to who should hear an appeal from the UT or to the effect that the other appellate court was wrong on something. Certainly less often that CA judges disagree with each other! However, it seems to me that Tehrani is essentially about the principles governing which court has judicial review jurisdiction in individual cases. It was not about whether a decision by the CS or the CA on GB-wide legislation binds tribunals in the other part of GB when dealing with the same point of law.
Para 24 of Tehrani: http://www.bailii.org/uk/cases/UKHL/2006/47.html
Note the earlier reference to Employment Tribunals being in the same position at 22.
It is exactly about whether a decision by the CoS or the CoA binds tribunals across the UK on GB wide legislation.
It would, I think, be entirely possibly for the E&W Court of Appeal to hear an appeal of a UT decision on the basis that the decision had failed to follow precedent established in the Court of Session.
Giles, I disagree that “Nelson effectively found that any ‘bedroom’ that was sufficient to accommodate anyone on the B13(5) list was a bedroom.” In fact it says at para 27 on this issue that:
(ii) designation or choices … as to to who should occupy rooms as bedrooms or how rooms should be used is unlikely to have an impact on the application of the regulation,” Which alludes to your comment yet it goes on to say underneath this at 27:
“We have not expressed point (iii) in absolute terms because it was not the focus of argument in this case and without such focused argument we do not consider that it would be appropriate to say any such designation or choice can never be relevant …”
This case – the connection issue as termed – is and was focused argument as to the central point of suitability of a room to accommodate two children of the same sex in any given room claimed to be a bedroom.
Nelson goes on at 35 to say “Issues as to whether a room of that size is a bedroom because it could be used as a bedroom for one child under 10, but not a teenager under the age of 16, are outside the ambit of this decision” and we must read that as meaning outside the decision as to whether a room is a bedroom if it can’t accommodate two children under 10 as was the central issue here.
[35] of Nelson goes on to say:
“…we note that paragraph 5 of bulletin U6/2013 and the Secretary of States submission to us seem to indicate that his view is that there must be room for a normal single bed and so if there was only room for say one cot or one young child’s bed he would not, or would not generally, agree the room as a bedroom.”
Hence it for me is flawed to say “Nelson effectively found that any ‘bedroom’ that was sufficient to accommodate anyone on the B13(5) list was a bedroom.” as that posit means an eclectic choice from B13(5) to mean three rooms accommodating only a small child is a 3 bed for bedroom tax purposes which this RH decision flatly rejected. Nelson did not give a lowest common denominator in terms of what is a bedroom
Joe, the Nelson decision is a complete mess, as we all know. It did find exactly what I said. But then went on to make various confusing remarks about issues that it wasn’t making a finding on but that may indicate otherwise. But this is rather by the by – the findings that it did make were sufficient for this UT to have to distinguish it and where there was not possible, decide otherwise. And, as another three judge UT, it can do exactly that, with no issues of precedent etc arising.
The Court of Session decision is the much, much bigger problem.
Imagine a case on exactly the same points coming to the First-tier Tribunal from Scotland. Are they bound by the Court of Session judgment or the newer Upper Tribunal one which stated it were not bound by the CoS when making its decision on a case originated from England?
Leaving aside the horribly tangled cross-border precedent issue, I do think it is possible to reconcile these three cases.
In both IB and Nelson there was a room of some kind in addition to the bedrooms used by the Reg 13(5) occupier categories. The issue was whether that room was a bedroom. The “vacant possession” remarks should be read in that context. The question whether any “spare” room was a bedroom only arose because the household members had enough undisputed bedrooms. The new case is looking at a completely different issue: whether there are enough bedrooms to accommodate the Reg 13(5) categories in the most efficient combination before we even consider the status of any “spare” rooms. So I think it is possible to follow all the decisions without inherent contradiction:
Stage 1 (the Nuneaton approach): distribute the occupiers as efficiently as possible under Reg 13(5) among the bedrooms – this process has to take into account how realistically usable the bedrooms are by the Reg 13(5) categories.
Stage 2 (the Nelson approach, slightly tempered by IB) Of the remaining rooms unallocated at stage 1, are there any that could realistically be used as a bedroom by anyone – not necessarily someone who currently lives here, but any prospective tenant viewing this property while it is vacant? If so, the bedroom tax applies.
That is a reasonable approach. I somehow doubt it will be the DWP’s interpretation of IB, though.